CN112418574A - Artificial intelligence-based urban rail transit operation simulation system and method - Google Patents
Artificial intelligence-based urban rail transit operation simulation system and method Download PDFInfo
- Publication number
- CN112418574A CN112418574A CN201910773966.1A CN201910773966A CN112418574A CN 112418574 A CN112418574 A CN 112418574A CN 201910773966 A CN201910773966 A CN 201910773966A CN 112418574 A CN112418574 A CN 112418574A
- Authority
- CN
- China
- Prior art keywords
- passenger
- information
- rail transit
- model
- intelligent agent
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004088 simulation Methods 0.000 title claims abstract description 146
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 60
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000006399 behavior Effects 0.000 claims description 40
- 238000012937 correction Methods 0.000 claims description 15
- 230000008859 change Effects 0.000 claims description 6
- 230000007704 transition Effects 0.000 claims description 6
- 230000000694 effects Effects 0.000 abstract description 9
- 238000004422 calculation algorithm Methods 0.000 description 32
- 238000004458 analytical method Methods 0.000 description 20
- 230000008569 process Effects 0.000 description 19
- 238000007726 management method Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 9
- 230000000007 visual effect Effects 0.000 description 5
- 238000013486 operation strategy Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000010485 coping Effects 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011158 quantitative evaluation Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention provides an artificial intelligence-based urban rail transit operation simulation system and method, which comprises the following steps: constructing a rail transit network model module: constructing a rail transit network model, and acquiring information of the constructed rail transit network model; a passenger classification module: applying AFC data to classify passengers, generating passenger agents, and acquiring passenger classification information and passenger agent information; a path selection module: carrying out path selection by using historical passenger flow data to obtain path selection information; and a simulation result output module: and outputting simulation result information according to the constructed rail transit network model information, the passenger classification information, the passenger intelligent agent information and the path selection information. The invention increases the use effect of the operation simulation system and can better play a role in assisting decision.
Description
Technical Field
The invention relates to the field of artificial intelligence simulation, in particular to an artificial intelligence-based urban rail transit operation simulation system and method.
Background
The urban rail transit operation simulation system is an important basic tool in an urban rail transit passenger flow management system, and the application of the urban rail transit operation simulation system is divided into two aspects of posterior analysis and advance prediction. The post analysis is mainly based on scene restoration of historical data, and aims at the problem that the whole urban rail transit system cannot be monitored in a whole disc in reality, if partial real-time passenger flow data cannot be or is difficult to obtain, the scene restoration can be carried out by using simulation, operators are helped to carry out scene analysis, operation strategies are formulated, and particularly the determination of the train operation plan is carried out. The advance prediction is another main application aspect, and as urban rail transit often needs to face some large passenger flow scenes, including predetermined events or accidents, operators need to make reasonable coping methods; the simulation system provides a simulation platform, and can help operators to test the effect and the rationality of the coping method in advance, so that targeted transportation service is provided for passengers, and the riding experience is improved. The existing rail transit operation simulation system can be divided into two types, namely macroscopic simulation and microscopic simulation. The macro simulation aims at simulating the overall expression of the system, and independent participants are often replaced by data or models in the simulation without concerning the detailed expression of the independent participants and only show the law embodied macroscopically; microscopic simulation focuses more on each individual participant in the simulation object, and it is desirable to build the simulation system from bottom to top with the largest possible simulation of the participants. The macro simulation can realize the simulation function with smaller resource consumption, and the micro simulation can embody more details. With the development of computer hardware technology, microscopic simulation is becoming the first choice of urban rail transit simulation. In microscopic simulation, the technical difficulty lies in simulating the behavior of a single passenger, especially in a large-scale rail transit network, a plurality of reachable paths often exist between a start point and a stop point, and determining the path selection behavior of the passenger is a main obstacle for accurate simulation. The existing method is usually based on a utility function with a model to select a path, and part of the methods can classify passengers in advance and design different utility functions. However, the above method is usually based on the subjective thought of the designer, and does not necessarily correspond to the objective reality, and the simulation degree of the simulation is also negatively affected, thereby preventing further popularization and use of the urban rail transit operation simulation system.
Patent document CN108536965A provides a method for calculating reliability of operation service of an urban rail transit line, which relates to the technical field of urban rail train operation control, and the method determines the number of delayed trains and the delayed train delay time according to train operation simulation and a train operation diagram, thereby calculating the reliability of train right-point; meanwhile, according to the delay time, determining a line operation conveying capacity co-scheduling solving model, and calculating line operation conveying capacity co-scheduling and the position density of passengers in the train; and finally, according to the gain type weighted fusion model, combining the positive point reliability, the line operation transmission capacity co-scheduling and the position density, constructing an operation service reliability model and calculating the operation service reliability. However, the patent mainly takes train simulation as a main part and does not relate to passenger simulation, and the simulation degree still has space for improvement.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide an artificial intelligence-based urban rail transit operation simulation system and method.
The invention provides an artificial intelligence-based urban rail transit operation simulation system, which comprises: constructing a rail transit network model module: constructing a rail transit network model, and acquiring information of the constructed rail transit network model; a passenger classification module: applying AFC data to classify passengers, generating passenger agents, and acquiring passenger classification information and passenger agent information; a path selection module: carrying out path selection by using historical passenger flow data to obtain path selection information; and a simulation result output module: and outputting simulation result information according to the constructed rail transit network model information, the passenger classification information, the passenger intelligent agent information and the path selection information.
Preferably, the building of the rail transit network model module comprises: generating an agent model module: generating one or more intelligent agent models through a network structure of real rail transit; the agent model can operate independently.
Preferably, the generating an agent model module comprises: generating intelligent agent model modules in sequence: sequentially generating a station intelligent agent model, an interval intelligent agent model, a vehicle intelligent agent model and a passenger intelligent agent model through a network structure of real rail transit; the intelligent agent model can frequently interact with the similar intelligent agents and the non-similar intelligent agents, adjust self behaviors and reflect the change of the behaviors in a state transition mode; the different classes of agents have different characteristic parameter types and different behavior modes, and the agents of the same class are distinguished by characteristic parameter values.
Preferably, the passenger classification module includes: a passenger classification module: according to the information of passengers entering and leaving the station, an artificial intelligence algorithm is applied to classify the passengers, and classification result information is obtained; obtaining a diversified passenger modeling module: and combining the classification result with the passenger agent according to the classification result information, acquiring diversified passenger modeling by adjusting the characteristic parameters of the passenger agent, solidifying the passenger modeling in the passenger agent and using the passenger modeling under the scene without AFC data.
The passenger entrance and exit information includes any one or more of the following information: -position information of passengers entering and exiting the station; -time information of passengers entering and exiting the station. The artificial intelligence algorithm classifies passengers by analyzing data such as OD (origin-destination) and travel time of passengers, and has self-learning and unsupervised characteristics;
preferably, the path selection module comprises: a correct passenger routing results module: performing multiple times of simulation in the same real passenger flow data scene to obtain correction result information; a passenger routing module with corrected passenger routing results: according to the correction result information, applying an artificial intelligence algorithm to carry out passenger path selection in a mode of correcting the passenger path selection result; the artificial intelligence algorithm has the characteristics of self-learning and data driving.
The invention provides an artificial intelligence-based urban rail transit operation simulation method, which comprises the following steps: building a rail transit network model: constructing a rail transit network model, and acquiring information of the constructed rail transit network model; a passenger classification step: applying AFC data to classify passengers, generating passenger agents, and acquiring passenger classification information and passenger agent information; a path selection step: carrying out path selection by using historical passenger flow data to obtain path selection information; and (3) outputting a simulation result: and outputting simulation result information according to the constructed rail transit network model information, the passenger classification information, the passenger intelligent agent information and the path selection information.
Preferably, the step of constructing a rail transit network model comprises: generating an intelligent agent model: generating one or more intelligent agent models through a network structure of real rail transit; the agent model can operate independently.
Preferably, the generating of the agent model step comprises: sequentially generating an intelligent agent model: sequentially generating a station intelligent agent model, an interval intelligent agent model, a vehicle intelligent agent model and a passenger intelligent agent model through a network structure of real rail transit; the intelligent agent model can frequently interact with the similar intelligent agents and the non-similar intelligent agents, adjust self behaviors and reflect the change of the behaviors in a state transition mode; the different classes of agents have different characteristic parameter types and different behavior modes, and the agents of the same class are distinguished by characteristic parameter values.
Preferably, the passenger classifying and generating passenger agent step includes: a passenger classification step: according to the information of passengers entering and leaving the station, an artificial intelligence algorithm is applied to classify the passengers, and classification result information is obtained; acquiring diversified passenger modeling steps: according to the classification result information, the classification result is combined with the passenger agent, and diversified passenger modeling is obtained by adjusting the characteristic parameters of the passenger agent; and can be solidified inside the passenger intelligent agent and used in the scene without AFC data. The passenger entrance and exit information includes any one or more of the following information: -position information of passengers entering and exiting the station; passenger on-off time information, and an algorithm for classifying passengers by analyzing data such as passenger OD, travel time and the like, wherein the artificial intelligence algorithm has a self-learning and unsupervised characteristic.
Preferably, the path selecting step includes: correcting the passenger routing result: performing multiple times of simulation in the same real passenger flow data scene to obtain correction result information; a passenger routing step is carried out according to the corrected passenger routing result: according to the correction result information, applying an artificial intelligence algorithm to carry out passenger path selection in a mode of correcting the passenger path selection result; the artificial intelligence algorithm has the characteristics of self-learning and data driving; the artificial intelligence algorithm has the characteristics of self-learning and data driving.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention provides a highly simulated simulation platform for urban rail transit operation management personnel, and a user can verify operation strategies, analyze historical events afterwards, predict scenes and the like on the platform, thereby having the function of assisting decision-making;
2. the invention has wide application scenes, and suitable scenes comprise: the method comprises the following steps of sudden accident handling and practicing, driving planning, large passenger flow event handling, newly opened line verification and the like;
3. the invention utilizes the artificial intelligence algorithm to improve the simulation degree of the simulation model, increases the use effect of the operation simulation system, and can better play a role in assisting decision. At present, similar analysis tools are lacked in the rail transit operation management process, the system can be popularized to solve the problem, and the prospect is very wide.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a schematic flow chart of the present invention.
Fig. 2 is a schematic block diagram of an artificial intelligence-based urban rail transit operation simulation system provided by the invention.
Fig. 3 is a schematic diagram of the operation simulation system of the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
The invention provides an artificial intelligence-based urban rail transit operation simulation system, which comprises: constructing a rail transit network model module: constructing a rail transit network model by using a multi-agent microscopic simulation technology, and acquiring information of constructing the rail transit network model; a passenger classification module: carrying out passenger classification by applying AFC data through an artificial intelligence scheme, generating passenger agents, acquiring passenger classification and generating passenger agent information; a path selection module: carrying out path selection by applying historical passenger flow data through an artificial intelligence scheme, so that the passenger behavior is more consistent with an actual result, and path selection information is obtained; and a simulation result output module: and according to the information for constructing the rail transit network model and the passenger classification, generating passenger intelligent agent information and path selection information, and outputting simulation result information.
The invention aims to solve the technical problem of providing an artificial intelligence scheme-based urban rail transit operation simulation system. According to the scheme, a microscopic simulation framework and an artificial intelligence technology are adopted, and the problem that the behavior of passengers in an operation simulation model is accurately matched with a real scene is solved. Firstly, a rail transit simulation model is built by using a multi-intelligent microscopic simulation technology; then, generating different passenger agents by using historical data of the ticket selling and checking system; and finally, in the simulation process, reading historical passenger flow data by using an artificial intelligence scheme, and correcting the passenger behavior. The whole simulation process can be visualized.
Specifically, in one embodiment, an artificial intelligence-based urban rail transit operation simulation system includes: a microscopic simulation framework module; an artificial intelligence based passenger classification module; an artificial intelligence based passenger routing module.
A microscopic simulation framework module: reading parameters to generate a microscopic simulation framework, and generating a station intelligent agent model, an interval intelligent agent model, a vehicle intelligent agent model, a passenger intelligent agent model and the like. The intelligent agents are abstracted by extracting characteristic parameters by taking an object of a real scene as a prototype, and each intelligent agent acts independently according to a corresponding rule, can interact with the same type of intelligent agents and different types of intelligent agents and influences self behaviors.
Passenger classification module based on artificial intelligence: the passengers are classified without supervision according to information such as travel events, starting points and stopping points and the like by reading historical data provided by the ticket selling and checking system. And combining the classification result with the passenger intelligent body, and generating a passenger intelligent body model according with a real scene in a parameter adjustment mode.
Passenger routing module based on artificial intelligence: the scheme gives a default path to the passenger intelligent agent during the first simulation operation, and through multiple times of simulation, the path of the passenger can be corrected by reading historical passenger flow data, and a path which is more matched with a real result is selected. The corrected passenger routing behavior may be solidified inside the passenger agent as a default path for use by the simulation model.
The invention adopts a data-driven artificial intelligence scheme, and realizes an intelligent passenger modeling process on a microscopic simulation framework. The method comprises the steps of firstly reading external data through a parameter management system to generate a rail transit network model, secondly establishing a passenger intelligent agent by using a classification result, secondly generating a passenger path matched with a real scene by using a path selection scheme, and finally outputting a simulation result and carrying out primary analysis for a user to use.
Preferably, the building of the rail transit network model module comprises: generating an agent model module: generating one or more intelligent agent models through a network structure of real rail transit; the agent model can operate independently.
Preferably, the generating an agent model module comprises: generating intelligent agent model modules in sequence: sequentially generating a station intelligent agent model, an interval intelligent agent model, a vehicle intelligent agent model and a passenger intelligent agent model through a network structure of real rail transit; the intelligent agent model can frequently interact with the similar intelligent agents and the non-similar intelligent agents, adjust self behaviors and reflect the change of the behaviors in a state transition mode; the different classes of agents have different characteristic parameter types and different behavior modes, and the agents of the same class are distinguished by characteristic parameter values.
Preferably, the passenger classification module includes: a passenger classification module: according to the information of passengers entering and leaving the station, an artificial intelligence scheme is applied to classify the passengers, and classification result information is obtained; obtaining a diversified passenger modeling module: and combining the classification result with the passenger agent according to the classification result information, acquiring diversified passenger modeling by adjusting the characteristic parameters of the passenger agent, solidifying the passenger modeling in the passenger agent and using the passenger modeling under the scene without AFC data.
The passenger entrance and exit information includes any one or more of the following information: -position information of passengers entering and exiting the station; -time information of passengers entering and exiting the station. The artificial intelligence scheme classifies passengers by analyzing data such as OD (origin-destination) and travel time of the passengers, and has self-learning and unsupervised characteristics.
Preferably, the path selection module comprises: a correct passenger routing results module: performing multiple times of simulation in the same real passenger flow data scene to obtain correction result information; a passenger routing module with corrected passenger routing results: and according to the correction result information, applying an artificial intelligence scheme to perform passenger routing in a manner of correcting the passenger routing result.
Specifically, in one embodiment, the correct passenger routing results module includes: firstly, the scheme gives a default path to passengers, real passenger flow indexes are compared after simulation is finished, and the path selection result of the passengers is gradually adjusted to be more in line with real data. The whole adjusting and optimizing process is independently realized by a scheme. The passenger routing result after the scheme correction can be solidified in the passenger intelligent body and directly used as a default value of routing so as to deal with simulation under the scene of non-real data or real historical data. The whole simulation process will be presented in a visual manner.
Specifically, in one embodiment, the artificial intelligence-based urban rail transit operation simulation system comprises an input management system module, a wire network simulation model module, a passenger path generation scheme module, a passenger behavior analysis module and a simulation result analysis module. In the system operation process, an input management system firstly reads OD passenger flow data and simulation parameters and inputs a simulation model; meanwhile, the passenger behavior analysis scheme reads Automatic Fare Collection (AFC) data, classifies passengers according to preset characteristics, and returns results to the simulation model; before the simulation model is started, a passenger path generation scheme generates a set of initial paths according to the characteristics of each simulated passenger; after the model is started, the behaviors and the interaction processes of passengers, vehicles and facilities in the rail transit system can be simulated and displayed in a visual mode; after the simulation is finished, the result analysis system can carry out quantitative evaluation on the simulation process and give corresponding indexes, meanwhile, the passenger behavior analysis scheme can automatically adjust the initial path scheme according to the difference of the corresponding indexes and the real result, and finally, the simulation effect with higher precision is realized through multiple times of simulation. The simulation model is based on a multi-agent microscopic simulation technology, and can realize the simulation of single passenger, vehicle or station facilities; the passenger behavior analysis and passenger path generation scheme is based on an artificial intelligence scheme with self-learning capability, a passenger intelligent agent model is established by repeatedly analyzing and calculating real passenger flow data, and an intelligent agent is endowed with a behavior mode similar to that of a real passenger, so that the simulation degree is improved. Compared with the existing urban rail transit operation simulation system, the scheme adopts a microscopic simulation framework, and can embody more detailed parameters than macroscopic simulation; meanwhile, a data-driven artificial intelligence scheme is used in a passenger behavior analysis and path generation module, real historical passenger flow data are utilized, the operation details of the urban rail system in reality are restored in a result reverse-pushing process, the result is solidified in a passenger intelligent body, and the reliability of a simulation effect can be greatly improved no matter real passenger flow data or virtual passenger flow data are adopted.
The invention provides an artificial intelligence-based urban rail transit operation simulation method, which comprises the following steps: building a rail transit network model: constructing a rail transit network model by using a multi-agent microscopic simulation technology, and acquiring information of constructing the rail transit network model; a passenger classification step: the method comprises the steps of applying AFC data to classify passengers through an artificial intelligence algorithm, generating passenger agents, and obtaining passenger classification information and passenger agent information; a path selection step: carrying out path selection by applying historical passenger flow data through an artificial intelligence algorithm, so that the passenger behavior is more consistent with an actual result, and path selection information is obtained; and (3) outputting a simulation result: and outputting simulation result information according to the constructed rail transit network model information, the passenger classification information, the passenger intelligent agent information and the path selection information.
Preferably, the step of constructing a rail transit network model comprises: generating an intelligent agent model: generating one or more intelligent agent models through a network structure of real rail transit; the agent model can operate independently.
Preferably, the generating of the agent model step comprises: sequentially generating an intelligent agent model: sequentially generating a station intelligent agent model, an interval intelligent agent model, a vehicle intelligent agent model and a passenger intelligent agent model through a network structure of real rail transit; the intelligent agent model can frequently interact with the similar intelligent agents and the non-similar intelligent agents, adjust self behaviors and reflect the change of the behaviors in a state transition mode; the different classes of agents have different characteristic parameter types and different behavior modes, and the agents of the same class are distinguished by characteristic parameter values.
Preferably, the passenger classifying and generating passenger agent step includes: a passenger classification step: according to the information of passengers entering and leaving the station, an artificial intelligence algorithm is applied to classify the passengers, and classification result information is obtained; acquiring diversified passenger modeling steps: according to the classification result information, the classification result is combined with the passenger agent, and diversified passenger modeling is obtained by adjusting the characteristic parameters of the passenger agent; and can be solidified inside the passenger intelligent agent and used in the scene without AFC data.
The passenger entrance and exit information includes any one or more of the following information: -position information of passengers entering and exiting the station; passenger on-off time information, and an algorithm for classifying passengers by analyzing data such as passenger OD (origin-destination) and travel time, wherein the artificial intelligence algorithm has a self-learning and unsupervised characteristic;
preferably, the path selecting step includes: correcting the passenger routing result: performing multiple times of simulation in the same real passenger flow data scene to obtain correction result information; a passenger routing step is carried out according to the corrected passenger routing result: according to the correction result information, applying an artificial intelligence algorithm to carry out passenger path selection in a mode of correcting the passenger path selection result; the artificial intelligence algorithm has the characteristics of self-learning and data driving.
Specifically, in one embodiment, the step of correcting the passenger routing results comprises: firstly, the algorithm gives the passengers a default path, the real passenger flow indexes are compared after the simulation is finished, and the path selection results of the passengers are gradually adjusted to be more in line with the real data. The whole adjusting and optimizing process is realized by an algorithm independently. The passenger routing result after algorithm correction can be solidified in the passenger intelligent body and directly used as a default value of routing so as to deal with simulation under the scene of non-real data or real historical data. The whole simulation process will be presented in a visual manner.
Specifically, in one embodiment, the artificial intelligence-based urban rail transit operation simulation method comprises an input management system, a wire network simulation model, a passenger route generation algorithm, a passenger behavior analysis algorithm and a simulation result analysis system. In the system operation process, an input management system firstly reads OD passenger flow data and simulation parameters and inputs a simulation model; meanwhile, the passenger behavior analysis algorithm reads Automatic Fare Collection (AFC) data, classifies passengers according to preset characteristics, and returns results to the simulation model; before the simulation model is started, a passenger path generation algorithm generates a set of initial paths according to the characteristics of each simulated passenger; after the model is started, the behaviors and the interaction processes of passengers, vehicles and facilities in the rail transit system can be simulated and displayed in a visual mode; after the simulation is finished, the result analysis system can carry out quantitative evaluation on the simulation process to give corresponding indexes, meanwhile, the passenger behavior analysis algorithm can automatically adjust an initial path scheme according to the difference of the corresponding indexes and the real result, and finally, a simulation effect with higher precision is realized through multiple times of simulation. The simulation model is based on a multi-agent microscopic simulation technology, and can realize the simulation of single passenger, vehicle or station facilities; the passenger behavior analysis and passenger path generation algorithm is based on an artificial intelligence algorithm with self-learning capability, a passenger intelligent agent model is established by repeatedly analyzing and calculating real passenger flow data, and an intelligent agent is endowed with a behavior mode similar to that of a real passenger, so that the simulation degree is improved. Compared with the existing urban rail transit operation simulation system, the scheme adopts a microscopic simulation framework, and can embody more detailed parameters than macroscopic simulation; meanwhile, a data-driven artificial intelligence algorithm is used in a passenger behavior analysis and path generation module, real historical passenger flow data are utilized, the operation details of the urban rail system in reality are restored in a result reverse-pushing process, the result is solidified in a passenger intelligent body, and the reliability of a simulation effect can be greatly improved no matter real passenger flow data or virtual passenger flow data are adopted.
The invention provides a data-driven passenger intelligent agent model generation method by applying a multi-intelligent agent and artificial intelligence technology, and is applied to an urban rail transit model. In the architecture, a multi-agent micro simulation method is adopted to achieve the purpose of reflecting more application details. In the algorithm, a passenger intelligent body is constructed by utilizing an artificial intelligence technology, and the characteristics of passengers are summarized and summarized by adopting a classification algorithm to simulate passengers with different requirements and individuality; secondly, a self-learning algorithm is used for generating passenger paths, and real passenger flow data are used for correction in the simulation process, so that a better simulation effect is achieved. Finally, the simulation operation process and the evaluation indexes are displayed through visual simulation, and urban rail transit operators are helped to make operation strategies, such as determining a train operation plan and a station pedestrian flow guiding strategy.
The invention provides a highly simulated simulation platform for urban rail transit operation management personnel, and a user can verify operation strategies, analyze historical events afterwards, predict scenes and the like on the platform, thereby having the function of assisting decision-making; the invention has wide application scenes, and suitable scenes comprise: the method comprises the following steps of sudden accident handling and practicing, driving planning, large passenger flow event handling, newly opened line verification and the like; the invention utilizes the artificial intelligence algorithm to improve the simulation degree of the simulation model, increases the use effect of the operation simulation system, and can better play a role in assisting decision. At present, similar analysis tools are lacked in the rail transit operation management process, the system can be popularized to solve the problem, and the prospect is very wide.
Those skilled in the art will appreciate that, in addition to implementing the system and its various devices, modules, units provided by the present invention as pure computer readable program code, the system and its various devices, modules, units provided by the present invention can be fully implemented by logically programming method steps in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system and various devices, modules and units thereof provided by the invention can be regarded as a hardware component, and the devices, modules and units included in the system for realizing various functions can also be regarded as structures in the hardware component; means, modules, units for performing the various functions may also be regarded as structures within both software modules and hardware components for performing the method.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (10)
1. The utility model provides an urban rail transit operation simulation system based on artificial intelligence which characterized in that includes:
constructing a rail transit network model module: constructing a rail transit network model, and acquiring information of the constructed rail transit network model;
a passenger classification module: applying AFC data to classify passengers, generating passenger agents, and acquiring passenger classification information and passenger agent information;
a path selection module: carrying out path selection by using historical passenger flow data to obtain path selection information;
and a simulation result output module: and outputting simulation result information according to the constructed rail transit network model information, the passenger classification information, the passenger intelligent agent information and the path selection information.
2. The artificial intelligence-based urban rail transit operation simulation system according to claim 1, wherein constructing a rail transit network model module comprises:
generating an agent model module: generating one or more intelligent agent models through a network structure of real rail transit;
the agent model can operate independently.
3. The artificial intelligence based urban rail transit operation simulation system according to claim 2, wherein generating an intelligent agent model module comprises:
generating intelligent agent model modules in sequence: sequentially generating several intelligent agent models, namely a station intelligent agent model, an interval intelligent agent model, a vehicle intelligent agent model and a passenger intelligent agent model, through a network structure of the real rail transit;
the intelligent agent model can interact with the same-class intelligent agents and non-same-class intelligent agents, adjust self behaviors and embody the change of the behaviors in a state transition mode;
the different classes of agents have different characteristic parameter types and different behavior modes, and the agents of the same class are distinguished by characteristic parameter values.
4. The artificial intelligence based urban rail transit operation simulation system according to claim 1, wherein the passenger classification module comprises:
a passenger classification module: classifying passengers according to the information of passengers entering and leaving the station to obtain classification result information;
obtaining a diversified passenger modeling module: according to the classification result information, the classification result is combined with the passenger agent, and diversified passenger modeling is obtained by adjusting the characteristic parameters of the passenger agent;
the passenger entrance and exit information includes any one or more of the following information:
-position information of passengers entering and exiting the station;
-time information of passengers entering and exiting the station.
5. The artificial intelligence based urban rail transit operation simulation system according to claim 1, wherein the path selection module comprises:
a correct passenger routing results module: performing multiple times of simulation in the same real passenger flow data scene to obtain correction result information;
a passenger routing module with corrected passenger routing results: and according to the correction result information, carrying out passenger routing in a mode of correcting the passenger routing result.
6. An artificial intelligence-based urban rail transit operation simulation method is characterized by comprising the following steps:
building a rail transit network model: constructing a rail transit network model, and acquiring information of the constructed rail transit network model;
a passenger classification step: applying AFC data to classify passengers, generating passenger agents, and acquiring passenger classification information and passenger agent information;
a path selection step: carrying out path selection by using historical passenger flow data to obtain path selection information;
and (3) outputting a simulation result: and outputting simulation result information according to the constructed rail transit network model information, the passenger classification information, the passenger intelligent agent information and the path selection information.
7. The artificial intelligence-based urban rail transit operation simulation method according to claim 6, wherein the step of constructing a rail transit network model comprises:
generating an intelligent agent model: generating one or more intelligent agent models through a network structure of real rail transit;
the agent model can operate independently.
8. The artificial intelligence based urban rail transit operation simulation method according to claim 7, wherein the generating an intelligent agent model step comprises:
sequentially generating an intelligent agent model: sequentially generating several intelligent agent models, namely a station intelligent agent model, an interval intelligent agent model, a vehicle intelligent agent model and a passenger intelligent agent model, through a network structure of the real rail transit;
the intelligent agent model can interact with the same-class intelligent agents and non-same-class intelligent agents, adjust self behaviors and embody the change of the behaviors in a state transition mode;
the different classes of agents have different characteristic parameter types and different behavior modes, and the agents of the same class are distinguished by characteristic parameter values.
9. The artificial intelligence based urban rail transit operation simulation method according to claim 6, wherein the passenger classification and generation step comprises:
a passenger classification step: classifying passengers according to the information of passengers entering and leaving the station to obtain classification result information;
acquiring diversified passenger modeling steps: according to the classification result information, the classification result is combined with the passenger agent, and diversified passenger modeling is obtained by adjusting the characteristic parameters of the passenger agent;
the passenger entrance and exit information includes any one or more of the following information:
-position information of passengers entering and exiting the station;
-time information of passengers entering and exiting the station.
10. The artificial intelligence-based urban rail transit operation simulation method according to claim 6, wherein the path selection step comprises:
correcting the passenger routing result: performing multiple times of simulation in the same real passenger flow data scene to obtain correction result information;
a passenger routing step is carried out according to the corrected passenger routing result: and according to the correction result information, carrying out passenger routing in a mode of correcting the passenger routing result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910773966.1A CN112418574A (en) | 2019-08-21 | 2019-08-21 | Artificial intelligence-based urban rail transit operation simulation system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910773966.1A CN112418574A (en) | 2019-08-21 | 2019-08-21 | Artificial intelligence-based urban rail transit operation simulation system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112418574A true CN112418574A (en) | 2021-02-26 |
Family
ID=74779718
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910773966.1A Pending CN112418574A (en) | 2019-08-21 | 2019-08-21 | Artificial intelligence-based urban rail transit operation simulation system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112418574A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112949081A (en) * | 2021-03-22 | 2021-06-11 | 中铁信(北京)网络技术研究院有限公司 | Rail transit road network emergency rescue strategy simulation and evaluation system |
| CN115203982A (en) * | 2022-09-14 | 2022-10-18 | 环球数科集团有限公司 | Parallel computing method and simulation system for intelligent operation of public transport vehicle |
| CN116702520A (en) * | 2023-08-07 | 2023-09-05 | 北京交通发展研究院 | Passenger rail transit travel condition simulation method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927591A (en) * | 2014-03-24 | 2014-07-16 | 北京交通大学 | Urban rail transit emergency evacuation optimization method and system |
| CN104616076A (en) * | 2015-01-30 | 2015-05-13 | 北京交通大学 | Method and system for optimizing multi-line collaborative operation scheme of urban rail transit |
| CN105224999A (en) * | 2015-09-10 | 2016-01-06 | 北京市交通信息中心 | The real-time passenger flow forecasting of urban track traffic based on AFC data and system |
| CN107273999A (en) * | 2017-04-27 | 2017-10-20 | 北京交通大学 | A kind of Flow Prediction in Urban Mass Transit method under accident |
| CN109409560A (en) * | 2018-08-16 | 2019-03-01 | 北京交通大学 | Urban track traffic for passenger flow abductive approach based on Multi-Agent simulation |
-
2019
- 2019-08-21 CN CN201910773966.1A patent/CN112418574A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927591A (en) * | 2014-03-24 | 2014-07-16 | 北京交通大学 | Urban rail transit emergency evacuation optimization method and system |
| CN104616076A (en) * | 2015-01-30 | 2015-05-13 | 北京交通大学 | Method and system for optimizing multi-line collaborative operation scheme of urban rail transit |
| CN105224999A (en) * | 2015-09-10 | 2016-01-06 | 北京市交通信息中心 | The real-time passenger flow forecasting of urban track traffic based on AFC data and system |
| CN107273999A (en) * | 2017-04-27 | 2017-10-20 | 北京交通大学 | A kind of Flow Prediction in Urban Mass Transit method under accident |
| CN109409560A (en) * | 2018-08-16 | 2019-03-01 | 北京交通大学 | Urban track traffic for passenger flow abductive approach based on Multi-Agent simulation |
Non-Patent Citations (2)
| Title |
|---|
| 姚向明;赵鹏;乔珂;: "城市轨道交通网络客流分布仿真模型研究", 交通运输系统工程与信息, no. 06, pages 52 - 59 * |
| 董皓;康会峰;廖琦;李璇;: "基于多智能体的城市轨道交通列车延误仿真模型及应用", 都市快轨交通, no. 03, pages 42 - 46 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112949081A (en) * | 2021-03-22 | 2021-06-11 | 中铁信(北京)网络技术研究院有限公司 | Rail transit road network emergency rescue strategy simulation and evaluation system |
| CN112949081B (en) * | 2021-03-22 | 2024-03-22 | 中铁信(北京)网络技术研究院有限公司 | Rail transit network emergency rescue strategy simulation and evaluation system |
| CN115203982A (en) * | 2022-09-14 | 2022-10-18 | 环球数科集团有限公司 | Parallel computing method and simulation system for intelligent operation of public transport vehicle |
| CN116702520A (en) * | 2023-08-07 | 2023-09-05 | 北京交通发展研究院 | Passenger rail transit travel condition simulation method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103593535B (en) | Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration | |
| CN111814308B (en) | Acceleration test system for automatic driving system | |
| CN112418574A (en) | Artificial intelligence-based urban rail transit operation simulation system and method | |
| CN106846818B (en) | road network dynamic traffic flow prediction method based on Simulink simulation | |
| CN111667092A (en) | Method and system for predicting short-term passenger flow of rail transit based on graph convolution neural network | |
| KR101815511B1 (en) | Framework for Traffic Simulation and Method for Simulation using Framework | |
| CN110162003B (en) | Driving active service system test verification platform and test method thereof | |
| Mahut et al. | Calibration and application of a simulation-based dynamic traffic assignment model | |
| Volodarets et al. | Optimization of vehicle operating conditions by using simulation modeling software | |
| CN115952692B (en) | Road traffic simulation method and device, storage medium and electronic equipment | |
| CN114692419A (en) | Digital twin modeling method and system for urban rail transit system | |
| Guerrero-Ibanez et al. | A policy-based multi-agent management approach for intelligent traffic-light control | |
| Miletić et al. | Comparison of two approaches for preemptive traffic light control | |
| CN114462233A (en) | Microscopic traffic simulation method, computer device and storage medium | |
| Postorino et al. | Agents meet traffic simulation, control and management: A review of selected recent contributions | |
| Wenl et al. | LimSim: A long-term interactive multi-scenario traffic simulator | |
| Lansdowne | Traffic simulation using agent-based modelling | |
| Medrano-Berumen et al. | Development of a validation regime for an autonomous campus shuttle | |
| Rahbari | Help the genetic algorithm to minimize the urban traffic on intersections | |
| Nüñez et al. | Simulating railway and metropolitan rail networks: From planning to on-line control | |
| Che et al. | An open vehicle-in-the-loop test method for autonomous vehicle | |
| CN113343422A (en) | Rail transit operation simulation method and system | |
| Hanzl et al. | Simulation of Autonomous Mobility of Connected Vehicles in the Context of Real Conditions–a Case Study | |
| Dimanche et al. | Massive railway operating data visualization; a tool for ratp operating expert | |
| Pariota et al. | Motivating the need for an integrated software architecture for connected and automated vehicles technologies development and testing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |